The Tenth North American Caribou Workshop, Girdwood, Alaska, USA, 4-6 May, 2004.

Conservation of based on a recalculation of the 1961 aerial survey on the ,

Frank L. Miller, Samuel J. Barry & Wendy A. Calvert

Canadian Wildlife Service, Environment Canada, Prairie & Northern Region, Room 200, 4999 — 98th Avenue, Edmonton, Alberta T6B 2X3, Canada (corresponding author: [email protected]).

Abstract: The estimate of 25 845 Peary caribou (Rangifer tarandus pearyi) on the Queen Elizabeth Islands (QEI) in the Canadian High Arctic in summer 1961 is the only nearly range-wide 'benchmark' for the past number of caribou. No variances or confidence intervals were calculated for this estimate and no estimates were calculated for Peary caribou on the three major islands of Ellesmere, Devon, and Axel Heiberg. We reexamined the 1961 raw data by grouping the QEI into five island-complexes ('eco-units') and calculating, for each unit, the estimated number of caribou and the standard error, and the 95% confidence interval of the estimate, using a 'bootstrap' technique with 100 000 replications. Our goal was to provide an ecological basis for evaluating subsequent changes in numbers rather than relying on single-island evaluations. Our bootstrap reanalysis produced an estimate of 28 288 ± 2205 SE with a 95% CI of 20 436—37 031 Peary caribou on the QEI in summer 1961. Substantial differences in density were apparent among the five eco-units, with about a 50-fold difference from 0.01 caribou • km-2 in the Eastern eco-unit to 0.5 caribou • km-2 in the Northwestern eco-unit. The 1961 findings, with our subsequent reexamination, are crucial to any evaluation of trends for the number of Peary caribou on the QEI and the relative importance of individual eco-units for these animals. These findings also allow a more accurate evaluation of the magnitude of the subsequent decline of Peary caribou on the QEI during the last four decades and may help predict future potential levels for caribou in each of the five eco-units.

Key words: Canadian High Arctic Islands, conservation, distribution, management, numbers, Rangifer tarandus pearyi.

Rangifer, Special Issue No. 16: 65-75

Introduction Tener (1963) estimated there were 25 845 Peary ite multi-island caribou surveys were conducted in caribou (Rangifer tarandus pearyi) on the Queen WQEI with adequate temporal separation (Miller et Elizabeth Islands (QEI) in summer 1961. This initial, al., 1977a; Miller, 1987a, 1987b, 1988, 1989; Gunn & range-wide, systematic aerial survey of Peary caribou Dragon, 2002). There have been no range-wide sys¬ has never been repeated. Tener found that the dis¬ tematic aerial surveys for Peary caribou on the EQEI tribution of caribou among the QEI was markedly or on any major portion of that region since 1961. uneven: based on island sizes he used, 94% (24 363) Only parts of were subsequently were on western QEI (WQEI), which makes up 24% aerially surveyed (Riewe, 1973; Case & Ellsworth, of the total island landmass, and the remaining 6% 1991; Gauthier, 1996; Manseau et al., 2004). (1482) were on eastern QEI (EQEI), with 76% of the The 1961 survey of the three largest islands— landmass (Fig. 1). Ellesmere, Devon and Axel Heiberg—was hindered Before 1961, the only information on distribution by persistent bad weather in some sections (Tener, and numbers of Peary caribou across the QEI came 1963). Survey coverage was low and fragmented. As from a few interested geologists in the mid- and late a result, Tener (1963) did not quantitatively estimate 1950s. After Tener's 1961 survey, only three compos- the total number of caribou on those islands and his

Rangifer, Special Issue No. 16, 2005 65 "intuitive guesses" for them were exceedingly small. etation differences (Edlund & Alt, 1989; Bliss, The mean overall density of 0.003 caribou • km-2 was 1990; Edlund, 1990). The collective landmass of but a tiny fraction of what the usable range on those the relatively large islands within the five eco-units three islands could support (62% of the total usable is 416 740 km2. The portion of the 1961 survey range on the QEI occurs on these three islands). area that we judge to be usable by Peary caribou is Subsequent development of aerial survey tech¬ 301 978 km2 or about 72% of the entire QEI. The niques and statistical analyses allow us to reevaluate 2092 small islands (defined as each having an area the 1961 data to determine standard errors and con¬ <137 km2) with a total landmass of 2321 km2 are not fidence intervals. We emphasize that the application included in any calculations for the five eco-units. of these statistical methods in no way betters Tener's A few caribou do live on these small islands: usu¬ original 1961 data set. The benefits of our approach ally only seasonally, but sometimes even year-round are: 1) it refines Tener's (1963) estimate of abun¬ (Miller, 1995a, 1997, 1998). dance; 2) it permits more detailed comparisons with subsequent aerial surveys; and 3) it identifies five 1961 Aerial survey eco-units to serve as the best standard for evaluating Macpherson's (1961) summary of wildlife observa¬ future changes in Peary caribou numbers, distribu¬ tions by geologists in the mid- and late 1950s was tions, and movements. Our statistical analyses may available as an internal report in 1960 and provided also satisfy those individuals who place an unfounded some factual information about the numbers and degree of reliance on probability testing and statisti¬ distributions of Peary caribou or other animals on the cal significance instead of evaluating results in terms QEI. However, Tener (1963: 8) believed that those of biological significance (e.g., Yoccoz, 1991; Johnson, surveys were "not detailed enough in their animal 1999; Steidl et al., 2000; Anderson et al., 2001). sightings to be of real value in selecting the method Our objective is to recalculate caribou numbers and of survey." Discussions with P. Larkin (University densities using recently available accurate measures of British Columbia) led Tener to choose system¬ of island size, to place confidence intervals on those atic sampling as the best design for obtaining both estimates using raw data in a bootstrap simulation, distributions and numbers. Although systematic and to provide an ecological basis for the conservation surveying on the Canadian has of Peary caribou on the QEI. become a method used by most biologists for aerial surveys of caribou and muskoxen (Ovibos moschatus), Tener (1963) did not apply any statistical analysis Materials and methods to his 1961 survey results because he and P. Larkin Animals and study area concluded that a useful statistical approach was not The Canadian High Arctic Islands form the northern available at the time. apex of the North American continent and collec¬ The survey transect flights in 1961 employed tively comprise a landmass of 419 061 km2 (Fig. 1). two survey crews, each operating with a Piper The QEI are those islands entirely north of ca. 74°N Super Cub aircraft. The planes were equipped with latitude. Although there are 2126 islands in total, large, low-pressure tires ('tundra tires') to allow only 34 of them each exceed 136 km2 in size and are them to take-off and land on unprepared ground. considered 'large' islands. The 22 islands surveyed When possible, the crews flew the surveys at about in 1961 differ in size from 412 km2 to 196 236 km2 152 m (500 ft) above ground level and 137 km • h-1 and total 410 765 km2 or 98% of the entire QEI. The (85 miles • h-1) air speed. The single observer in each largest is Ellesmere Island, the tenth largest island in plane sat directly behind the pilot and limited his the world. The 12 large islands not surveyed in 1961 observations to the right side of the aircraft. Transect had a collective landmass of 5975 km2. width was about 402 m (0.25 mi). On each island, We divided the QEI into western (WQEI) and the first transect was selected at random and drawn eastern (EQEI) zones (Fig. 1). We further subdi¬ on a survey map; additional transects were then vided the QEI into five island-complexes, which we drawn parallel to it at fixed intervals. termed 'eco-units' (Fig. 1). The two major zones and Aerial coverage varied by island and was small, the five eco-units are based on their relative differ¬ averaging 4% (see Tener, 1963: Table 1). The inter¬ ences in numbers of caribou estimated in 1961, and val between transects was determined by the chosen caribou numbers, distribution, and movements or sampling intensity. Tener (1963) based his survey migrations, as measured between 1961 and 1997 intensity on the belief that using a large number of (e.g., Tener, 1963; Miller et al., 1977a, 1977b; Miller, small samples was better than using only a few large 1990a, 1990b, 1998, 2002; Gunn & Dragon, 2002), ones and that a large number of systematically spaced climate differences (Maxwell, 1981, 1997), and veg- transects probably ensured adequate representation

66 Rangifer, Special Issue No. 16, 2005 maps, as the original data forms no longer exist (J. S. Tener, pers. comm., 2001). During his initial analysis, Tener plotted his observa¬ tions on the maps from the data forms, but after he completed the survey. We compared our raw data tabu¬ lations from those maps to the text, tables and figures in Tener (1963). To reconcile minor discrepancies between data on the maps and Tener (1963), Frank L. Miller and Samuel J. Barry each inde¬ pendently tabulated observa¬ tions of Peary caribou from each data map. They repeated this process four more times over the course of several Fig. 1. Queen Elizabeth Islands split into two major zones, western and eastern months, leaving sufficient Queen Elizabeth Islands; and five eco-units, Southwestern (SW), South-cen• time between repetitions to tral (SC), Northwestern (NW), North-central (NC), and Eastern (E): islands decrease 'remembering' their surveyed by Tener in 1961, (1) Melville, (2) Prince Patrick, (3) Eglinton, (4) Emerald, (5) Bathurst, (6) Cornwallis, (7) Lougheed, (8) Vanier, (9) Cameron, previous interpretation. They (10) Alexander, (11) Massey, (12) Little Cornwallis, (13) Mackenzie King, then did one joint count to (14) Borden, (15) Brock, (16) Ellef Ringnes, (17) Amund Ringnes, (18) confirm that their reasoning Cornwall, (19) King Christian, (20) Ellesmere, (21) Devon, and (22) Axel on each observation was the Heiberg. same. Finally, they compared the totals for each island and of range types and the geographical distribution of for each smaller survey land unit (stratum) used by the animals within them. Tener (1963) on those islands. Observers used previously prepared forms to record animals within the survey strip as being 'on-transect' Ca lc ulation of estim ated nu mbers a nd con fi- and those seen beyond as being 'off-transect.' Tener dence intervals (1963: 9) believed that the selected transect width We now know that many, perhaps most, caribou on and flight altitude facilitated the sighting of all ani¬ the QEI make seasonal inter-island migrations (e.g., mals on each transect and accurate identification of Miller et al., 1977a, 1977b; Miller, 1990a, 1995a, them by sex and age classes. The exception to this 2002). The objective of this evaluation of the 1961 was on parts of where, to take advan¬ aerial survey findings (Tener, 1963) was to recal¬ tage of the good weather while the Piper Super Cubs culate the estimates by grouping the 22 individual were not ready for service, he used a deHavilland islands in a more ecologically sound manner into five Beaver aircraft. They surveyed two transects each island-complexes and to attach a measure of vari¬ about 805 m (0.5 mi) wide by flying the Beaver air¬ ability and confidence to the estimates. At the level craft at about 305 m (1000 ft) above ground level. of our five eco-units, we used a 'bootstrap' procedure During the 70-d period between 10 June and 18 (Krebs, 1999) to estimate the number of caribou, August 1961, prevailing weather at the time permitted standard errors, and confidence intervals. We then survey flights on only 34 (49%) days. The Piper Super summed the eco-unit statistics to obtain values for Cubs flew about 500 h and the Beaver flew about WQEI, EQEI, and the entire QEI. 100 h, but only about 205 h were on survey transects. Our bootstrap procedure is as follows. Using only those observations of caribou seen on-transect, we Reexamination of 1961 data simulated a 'resurvey' of each island 100 000 times. Our approach On each resurvey we selected the same number of We relied on the observations plotted and tran¬ transects that Tener had flown in 1961 randomly sect lines drawn on the original 14 field survey and with replacement from the set of transects for

Rangifer, Special Issue No. 16, 2005 67 that island. For each of these 100 000 resurveys we bou in groups of two or more individuals constituted summed the number of caribou 'observed' and the 78% (306) of the observations and the remaining 87 area 'surveyed' from the randomly selected transects. (22%) were individual caribou. By our calculations, For each of these resurveys, we calculated the number Tener flew 27 872 km of transect lines and surveyed of caribou within each eco-unit by first pooling these 11 213 km2 (ca. 4%) of the usable range within the island results (caribou observed and area surveyed) 1961 survey area. The 100 000 estimates of caribou among the islands in the eco-unit then multiplying numbers formed by randomly resampling transect the resulting caribou density (total pooled caribou counts within each eco-unit yielded 28 288 ± 2205 observed divided by total pooled area surveyed) by the SE and a 95% CI of 20 436—37 031 Peary caribou on area of usable range within the eco-unit. Our bootstrap the QEI in summer 1961. Our estimate for the QEI estimate of caribou for each eco-unit is the mean of is nearly 9% greater than Tener's (1963) original esti¬ 100 000 of those estimates. We measured bootstrap mate (Table 2). Calves made up 20% of all caribou variability for each eco-unit as the variance of the observed throughout the QEI. bootstrap estimates (reported as standard error) and our bootstrap 95% confidence interval as the 2.5th and Two major zones 97.5th percentiles of the distribution of all bootstrap Our recalculation suggests that 90% of Peary cari¬ estimates of caribou numbers. Hereafter, 'caribou' is bou on the QEI occurred on the WQEI and only synonymous with 'estimated caribou' in all cases. 10% on the EQEI in summer 1961 (Table 2). The Although Tener (1963) used the best information number of caribou on WQEI is only 4% greater than available at the time for the sizes of the 22 islands the original number estimated there by Tener (Table he surveyed, the areas of those 22 islands have been 2). Tener (1963) did not calculate an estimate for the recalculated recently (http://atlas.gc.ca/english/facts/ islands surveyed in the Eastern eco-unit but used islands.html) and the reported sizes for 21 of them only "intuitive guesses"— collectively 650 caribou. have changed. Only , the When that 650 is combined with the 832 caribou smallest island surveyed in 1961, remained at its estimated by Tener (1963) on the four islands in the 1961 measured size of 412 km2. Therefore, we used North-central eco-unit there are 1482 caribou on the sizes for 13 of 21 islands that were smaller than the EQEI in 1961. Therefore, our estimate is 95% greater values used by Tener (1963) by 1% to 44% and for than the original 1961 collective island estimate the eight islands that were larger by 1% to 22%. plus guesses, although our 95% CI includes Tener's Thus, our estimates of numbers of caribou reflect 1961 combined estimate for the EQEI (Table 2). The these changes in island sizes, which cause some devi¬ WQEI and the EQEI represent 24% and 76% of ation from Tener's (1963) original 1961 estimates. the collective landmass of the 22 islands surveyed Ellesmere, Devon, and Axel Heiberg islands have in 1961 and 32% and 68% of the potential range large glaciated areas and high elevations that are for caribou, but the WQEI held nine times as many not suitable caribou range. Additionally, Melville caribou as the EQEI. Caribou exhibited a strong pref¬ Island has a small west-central permanent ice field. erence for WQEI with a density 18-fold greater than Tener (1963) excluded these areas from his calcula¬ that on the EQEI in summer 1961. tion of caribou numbers (but his calculations were based on different island sizes). We recalculated the Five eco-units size of permanently glaciated areas using GIS-based Southwestern 1:250 000 topographic maps with 150-m elevation There are five islands from 549 to 42 149 km2 in contours and the recently revised island sizes (Table size that make up this eco-unit (Table 1; Fig. 1). 3). As a result, we excluded all land > 750 m above However, (1150 km2) was not mean sea level (amsl) and the glaciated areas < 750 surveyed in 1961. Therefore, the eco-unit estimate is m amsl from the land area we used for computing based on the other four islands minus 159 km2 for ice estimates for the number of caribou present. This fields on western Melville Island (Table 1). There was removed all large areas of unsuitable range including a strong preference for Peary caribou in this eco-unit, all permanent snow and ice fields. We also excluded most (58%) of the caribou on the QEI and 64% of areas on Axel Heiberg and Ellesmere islands that those on WQEI occurred there (Table 2). received no aerial coverage. South-central There are 12 islands from 412 to 16 042 km2 in size Results that make up this eco-unit (Table 1; Fig. 1). The four We tallied 393 observations of 1583 caribou on-tran- smallest islands, Helena (326 km2), Baillie—Ham- sect on the QEI from Tener's 1961 field maps. Cari- ilton (290 km2), Griffith (189 km2), and Lowther 68 Rangifer, Special Issue No. 16, 2005 Table 1. Statistics compiled from 14 observation maps for the 1961 aerial survey of Peary caribou (Tener, 1963) and the most recent values for the 22 islands in the 1961 aerial survey, Queen Elizabeth Islands, Canadian Arctic Archipelago.

Total length Area surveyed Caribou observed Area used for calcu• Overall size of Eco-unit of transects (km) (km2) on-transect lating estimates (km2)a land unit (km2)b (Western QEI) (14 410) (S797) (1470) (96 067) (98 6S1) Southwestern 8163 3284 894 59 928 61 237 South-central 5221 2100 363 27 858 28 808 Northwestern 1026 413 213 8281 8606

(Eastern QEI) (13 462) (S416) (113) (194 535) (318 089) North-central 1929 776 30 19 553 19 SS3 Eastern 11 533 4640 83 174 982c 298 S36 a Includes only potentially usable range for Peary caribou. b Island sizes available from http://atlas.gc.ca/english/facts/islands.html. c Excludes 11 376 km2 of usable range that did not receive aerial coverage.

Table 2. The 1961 estimates of Peary caribou from Tener (1963) compared to the new estimates, associated standard errors, and confidence intervals calculated from the distribution of 100 000 bootstrap replications for each of five eco-units, Queen Elizabeth Islands, Canadian Arctic Archipelago. 1961 estimate 1961 esti¬ Bootstrap estimates of caribou Mean density of Eco-unit of Peary cari¬ mated % from this study estimated caribou

-2 bou calves Mean SE 95% CIb (caribou ' km ) (Western (24 363) (21) (286) (2S 401) (2109) (19 060-32 287) (0.264) QEI) Southwestern 1S 418 20 173 16 31S 1779 12 969-19 936 0.272 South-central 4933 21 89 4826 833 3283-6S38 0.173 Northwestern 4012 21 24 4260 767 2808-S813 0.S14

(Eastern (1482) (15) (208) (2887) (642) (1376-4744) (0.01S) QEI) North-central 832 21 38 760 293 269-140S 0.039

Eastern 6S0c 12 170 2127 S71 1107-3339 0.012 a N equals number of aerial survey transects flown. b 95% CI based on the 2.5th and 97.5th percentiles of 100 000 bootstrap estimates of caribou numbers. c Estimates for Ellesmere, Devon, and Axel Heiberg islands were not calculated in 1961 but based only on "intuitive guesses" by Tener (1963).

Table 3. Vertical relief divisions of the three mountainous islands of Ellesmere, Devon, and Axel Heiberg, within the Eastern eco-unit, eastern Queen Elizabeth Islands, Canadian Arctic Archipelago.

Size of % Area (km2) in 1S0-m bands % Area % Island island 0-1S0 151-300 301-450 451-600 601-750 > 750 of ice/ Usable (km2) snow range Ellesmere 196 236 10 13 15 13 12 37 43 57 Devon 55 247 12 19 15 38 5 11 20 80 Axel Heiberg 43 178 20 19 15 11 9 26 30 70 Totals 294 661 12 15 15 18 10 30 37 63

Rangifer, Special Issue No. 16, 2005 69 (145 km2) were not surveyed in 1961. Therefore, the this eco-unit was exceptionally low, based on usable eco-unit estimate is based on the collective landmass range for caribou (Table 2). Calves were appreciably of Bathurst, Cornwallis, Lougheed, Vanier, Cameron, lower in this eco-unit than in the other four eco-units Alexander, Massey, and Little Cornwallis islands (Table 1: %2 = 21.55, 4 df; P < 0.005). (Table 1: Tener's (1963) "Bathurst Island" included Bathurst and the islands of Vanier, Cameron, Alexan¬ der, and Massey). Caribou within this eco-unit made Discussion the second largest contribution to the estimate for the Tener was greatly limited in any interpretation or QEI (Table 2: 17%). evaluation of his results, with almost no on-site background information to draw from. In 1961, it Northwestern appeared that there was a relatively large number of There are three islands from 764 to 5048 km2 in Peary caribou available on the QEI to support the size that make up this eco-unit (Table 1; Fig. 1). establishment of settlements there. Unfortu¬ Thus, the eco-unit estimate is based on the collective nately, that condition has not persisted and Peary landmass of Mackenzie King, Borden, and Brock caribou on the QEI are now 'Endangered' (Gunn et islands, excluding 325 km2 from due al., 1981; Miller, 1990b; Miller & Gunn, 2003a, b). to fog (Table 1). Caribou within this eco-unit made Peary caribou populations on the WQEI now occur the third largest contribution to the estimate on the as mere remnants of their 1961 sizes (Tener, 1963; QEI (Table 2: 15%). Although the number of caribou Miller et al., 1977a; Miller, 1995b, 1998; Gunn et al., contributed from this eco-unit to the QEI estimate 2000; Gunn & Dragon, 2002). With the benefit of was only moderate, the density was exceptionally 4 decades of fragmentary information, we can extend high (Table 2). our assessment of the 1961 findings. In summer 1961 the relative mean densities and North-central distribution of Peary caribou populations on the QEI There are four islands from 645 to 11 295 km2 in indicated strong preferences for the Northwestern, size that make up this eco-unit (Table 1; Fig. 1). All Southwestern and South-central eco-units, low occur¬ four islands were surveyed in 1961: Ellef Ringnes rence in the North-central eco-unit, and extremely (11 295 km2), Amund Ringnes (5255 km2), Cornwall low occurrence in the Eastern eco-unit. Since then (2358 km2), and King Christian (645 km2). Caribou only the Southwestern and South-central eco-units within this eco-unit contributed the least to the have served as 'heartlands' for Peary caribou. QEI estimate (Table 2: 3%). The density of caribou within this eco-unit was markedly low (Table 2). Two major zones In summer 1961 there were 18-fold more Peary Eastern caribou per unit area of usable range on the WQEI There are 10 islands from 137 to 196 236 km2 in than on the EQEI (Fig. 1). Subsequent aerial sur¬ size that make up this eco-unit (Table 1; Fig. 1): veys suggested that the highest estimates of Peary Ellesmere (196 236 km2), Devon (55 247 km2), Axel caribou were still in the Southwestern and the South- Heiberg (43 178 km2), Graham (1378 km2), Meighen central eco-units (e.g., Miller et al., 1977a; Miller, (955 km2), North Kent (590 km2), Coburg (344 km2), 1990b; Gunn & Dragon, 2002). The high numbers Stor (313 km2), Hoved (158 km2), and Buckingham of caribou estimated in summer 1961 within the (137 km2). Only the three largest islands of Ellesmere, Northwestern eco-unit and on in Devon, and Axel Heiberg were surveyed in 1961. the South-central eco-unit have not been observed Aerial coverage was low and fragmented due to poor again (Miller et al., 1977a; Miller, 1990b; Gunn & weather. The potential range for caribou on the three Dragon, 2002). islands is 63% of their collective landmass because We believe the observed distribution of Peary 108 303 km2 are covered by glaciers (Table 3). Only caribou on the WQEI in 1961 reflected a summer¬ 27% of the land area on the three islands is <300 m time redistribution of the relatively high density of amsl; in contrast, about 90% of the entire WQEI lies caribou in the Southwestern eco-unit beyond the below the 300-m amsl level (Table 3; Miller et al., eco-unit boundaries to 'out islands' in the North¬ 1977a). More importantly, about 60% of the land¬ western eco-unit. Thus, the relatively high density mass within the Eastern eco-unit is at higher eleva¬ of caribou in the Northwestern eco-unit only reflects tions than those found anywhere else on the QEI. temporary ingress from elsewhere, not a persistent Caribou within this huge eco-unit composed only year-round state. It is also most likely that summer 7% of the total in the QEI, but 74% of the caribou range extended in 1961 to Lougheed Island, with in the EQEI (Table 2). The density of caribou within some spillover to the North-central eco-unit islands.

70 Rangifer, Special Issue No. 16, 2005 Whether this summertime redistribution involved This eco-unit is of particular importance because it caribou from both the Southwestern and South- is readily accessible by hunters. The Inuit settlement central eco-units is not clear, but Lougheed Island of Resolute Bay on Cornwallis Island was established appears to be more closely linked over time with in the mid-1950s. Bathurst Island became a primary changes in caribou numbers and distributions within hunting area and remained so until the major die-off the South-central eco-unit than with changes in the of caribou there in winter and spring 1973—74 (Parker Southwestern eco-unit. Lougheed Island could not et al., 1975; Miller et al., 1977a). Caribou abundance support caribou at a density of 1.1 • km-2 year-round had apparently been adequate to satisfy annual har¬ for more than a few years without overutilization of vest needs from at least the time of settlement until the range. The same applies to at 0.6 • 1973, when they were already decreasing in number. km-2 and Mackenzie at 0.5 • km-2, where Based on the past, this eco-unit could once again much of the range is on the poorly vegetated Beau¬ support a harvestable number of caribou, but prob¬ fort Formation (Tener, 1963; Edlund & Alt, 1989). ably only if the hunters from Resolute Bay give them However, surveys in winter have shown that sparsely time to recover. The three most recent annual die-offs vegetated but wind-blown range on the Beaufort between 1994 and 1997 saw the number of caribou formation can be important to the survival of Peary reduced from about 3000 to < 100 (Miller, 1998; caribou (Miller et al., 1977a). Gunn & Dragon, 2002; Miller & Gunn, 2003a).

Five eco-units Northwestern Southwestern This eco-unit is particularly remote from human After Tener (1963) completed his analyses, we learned settlement and the prevailing weather does not favor that many, if not most, caribou within this eco-unit systematic aerial surveys. Mackenzie King, Borden, make seasonal inter-island migrations, moving from and Brock islands all lie beyond the northern limit summer ranges on eastern Melville and Byam Mar¬ of prostrate shrubs, and sedges and vascular plants tin islands to winter ranges on Prince Patrick and are sparse, completely herbaceous, and least diverse of Eglinton islands (Miller et al., 1977a, 1977b). West¬ the QEI (Edlund & Alt, 1989; Edlund, 1990). This ern Melville Island appears to be important to Peary eco-unit has served on occasion as a 'spillover area' for caribou only during times of high numbers, such as caribou from the Southwestern eco-unit and possibly in summer 1961. The small size of Emerald Ile appar¬ the South-central eco-unit when the caribou on one ently relegates it to transitional range for some few or both eco-units uncommonly reached a relatively caribou, mainly during spring and autumn. high density. However, it is the smallest of the five Inter-island movements within this eco-unit must eco-units and there is no reason to believe that it have persisted over time, as Parry (1821) noted could sustain anything more than a relatively small Peary caribou moving off eastern Melville Island in low density 'reservoir area' for Peary caribou. autumn. Therefore, multi-island surveys of at least Melville and Prince Patrick islands are necessary to North-central estimate the number of caribou and any changes in This eco-unit lies mainly in the environmentally the number of caribou there over time. Ideally, Byam stressful northern zone of plant growth (Edlund & Martin, Eglinton, and Emerald islands should be Alt, 1989; Edlund, 1990). The ability of this range included in those surveys. to support a large number of caribou year-round is highly questionable given its moderate size and mea¬ South-central ger flora. In all likelihood, this eco-unit serves mainly Caribou within this eco-unit also make seasonal as a rarely used 'spillover area' for caribou from the inter-island migrations (Miller, 1995a, 1998, 2002); Southwestern eco-unit and possibly the South-central however, some remain year-round on Bathurst Island eco-unit when caribou on one or both eco-units infre¬ (Miller & Barry, 2003) and on other smaller islands quently reach relatively high densities. This eco-unit (Miller, 1997, 1998). The primary linkage appears should be viewed as a low density reservoir that can to be among Bathurst, Alexander, Marc, Massey, help foster the long-term persistence of Peary caribou Vanier, and Cameron islands. Ile Marc was not sur• on the QEI. The islands are not readily accessible to veyed in 1961: it is small (56 km2) but important Inuit hunters and could never provide any appreciable caribou summer range, lying between Alexander and level of sustained annual harvest. Massey islands. There is some evidence of movement between Bathurst and the other lesser satellite islands Eastern as well as Cornwallis and Lougheed islands (Miller, It is a mystery why there are not many more Peary 1997, 1998). caribou in this eco-unit. Its vastness makes Tener's

Rangifer, Special Issue No. 16, 2005 71 (1963) "intuitive guess" of only 650 Peary caribou where the annual level of harvest would not exceed there in 1961 appear ultra-conservative. This is 5% (cf. Kelsall, 1968), based on the mean estimate especially true as there are at least 186 000 km2 of minus 1 SE, minus the percentage of calves present, potential range within this eco-unit. Our estimate of the interval between surveys could be extended to 2127 ± 571 SE requires only 0.01 caribou • km-2 and 5 years. In all cases, however, an aerial survey still our estimate minus 1 SE (1556) would require, on should be carried out following any winter or spring average, only 0.008 caribou • km-2 of usable range. with exceptionally severe weather to prevent a major Based on the 1961 results and subsequent lim¬ decline in number going undetected. ited information, we view the Eastern eco-unit as The past and current status of Peary caribou in a huge, low-density reservoir for Peary caribou. In the EQEI remains debatable. No current information the foreseeable future, caribou there can make only exists for caribou in the North-central eco-unit. In a relatively small contribution to annual harvests the Eastern eco-unit, recent information from north¬ of Peary caribou on the QEI. However, because of ern Ellesmere Island suggests a stagnant condition the Eastern eco-unit's large and rugged land base, it with a minimum of only 45 Peary caribou estimated has great benefit to the persistence of Peary caribou there between 1988—2002 (Manseau et al., 2004). on the QEI. Also, given the paucity of data for the However, recent information from Inuit hunters from three major islands in this eco-unit and their large on southern Ellesmere Island suggests size, there is a chance that this area may have more that caribou are increasing there, based on sightings caribou than anyone is aware of. of a low number of caribou in places where they have Ironically, when the scenarios of 'global warming' been absent for several decades. Both M. Raillard, are considered, we could speculate that the Eastern Canadian Wildlife Service, and J. England, Univer¬ eco-unit may become the most important region for sity of Alberta, believe that while conducting other Peary caribou on the entire QEI. Because of the rug- field activities they have seen numbers of caribou gedness of its terrain and its elevation relief (Table on parts of Ellesmere and Axel Heiberg islands in 3), many micro niches will be available that could summers 1999—2001 that suggest increases are foster Peary caribou survival in that region compared probably occurring on some relatively well-vegetated to survival in the relatively low-lying WQEI and the sections. The accuracy of island-wide extrapolations North-central eco-unit of the EQEI. If global warm¬ of caribou numbers from such casual observations ing advances as predicted by some, this change could cannot be measured with confidence. Thus, such occur within the next century. Even more important¬ extrapolations form a poor basis for making decisions ly, it is possible that virtually only the Peary caribou regarding rates of annual harvest, particularly for an on the Eastern eco-unit may survive the transitional 'Endangered Species.' Only prescriptions for annual phase of greater snowfall and more frequent icing harvest that are based on actual counts of caribou events during global warming. The probability of that yield a total minimum number adequate to sus¬ this may be tied to the rapidity of warming processes tain desired harvests should be employed. Even then, which may cause micro niches to become critically following exceptionally severe weather years, harvest important to the caribou's long-term survival on the of Peary caribou may have to be temporarily halted QEI. by emergency action through cooperation with Inuit hunters. Management and conservation implications The vastness and concomitant prohibitive costs of Future aerial surveys of Peary caribou on the QEI aerial surveys on the Eastern eco-unit require a dif¬ should be at the eco-unit level to maximize both the ferent approach. We recommend the establishment accuracy of the estimated number of caribou and the of at least five permanent survey plots, 100 km by probability of detecting real changes in caribou num¬ 100 km on areas of known importance to Peary bers and distribution over time. Surveys should be caribou and where annual harvests are most likely flown in July, at three-year or shorter intervals as long to occur. These plots should be flown in early July as the number of caribou remains small and especial¬ when weather is most stable, and should be flown ly if they are being harvested annually. Regardless of each time as close as possible to the same days of this when the last survey was carried out, a July survey month to record the percentage of calves, and ide¬ should be flown in the summer immediately after a ally yearlings, present and the phenological state of winter or spring with exceptionally severe weather plants for comparisons among years. This procedure to determine the impact, if any, on the overwinter would provide a relative estimate of the number of survival of caribou and the production and early sur¬ caribou, calving success, yearling recruitment, and vival of calves. Once adequate aerial surveys indicate annual variation in new plant growth within those that the number of caribou has increased to a size 'food patches' that would allow the setting of flex-

72 Rangifer, Special Issue No. 16, 2005 ible maximum annual harvest levels. Accurate calf harvests had been at relatively high levels. Although and yearling counts should be made, as a measure there is no evidence that harvesting has caused Peary of performance to eventually provide a measure of caribou declines on the QEI, it could have accelerated their trend over time to better prescribe the maximal and deepened them in some instances. allowable level of annual caribou harvest, especially when annual harvest appears high. After the first year of plot surveys, both the previous number of Conclusions animals counted and the size of the desired future The Peary caribou on QEI is a unique part of annual harvest would dictate the needed frequency of Canada's High Arctic natural heritage and is recog¬ those aerial searches. In the absence of a better mea¬ nized as an 'Endangered' form of wildlife in Canada. sure, a maximum of a 5% annual harvest should be The number of caribou now on the QEI is a vestige adhered to, especially as long as Peary caribou remain of the number estimated there in 1961; yet, Peary an 'Endangered Species.' caribou continue to be considered a usable renewable It is most unlikely that the concepts of 'carrying resource! Our recalculation of the 1961 aerial survey capacity' and 'sustained yield' have any real applica¬ is the best benchmark for subsequent evaluations of tion to Peary caribou on the QEI. To date, everything changes in numbers, distributions, and movements of suggests that the caribou on the QEI function in a Peary caribou on the QEI. The division of the entire 'nonequilibrium grazing system' where abiotic fac¬ QEI into five eco-units provides a standard for evalu¬ tors, mainly snow and ice, control their fate through ating the above changes over time on an ecologically infrequently occurring, sporadic and unpredict¬ sound basis for prescribing and assessing conserva¬ able 'exceptionally hard weather years' (Caughley & tion measures for Peary caribou on the QEI. Gunn, 1993; Behnke, 2000; Miller & Gunn, 2003a, Continued annual harvesting of Peary caribou—an b). At such times, extremely unfavorable snow and 'Endangered Species'— requires that special safe¬ ice conditions prevent the animals from getting to guards and an adequate monitoring program be food or cause caribou to use more energy accessing established. Cooperation among wildlife agencies it than they recover in forage intake. In years when and Inuit hunters is essential, because precautionary such snow and ice conditions are prolonged and wide¬ measures will require the participation of Inuit users spread, a large number of caribou (and muskoxen) in setting the restrictions and complying with them. will die from starvation. This happens because the No more than a few Peary caribou should be taken carrying capacity of those caribou ranges will change, from any island within any eco-unit without first both quickly and unpredictably, with the prevailing obtaining actual counts to determine that enough severity of the snow and ice cover in each year. The caribou exist there to sustain the desired level of more severe the snow and ice conditions and the annual harvest. Otherwise, we run the risk of being longer they persist, the greater the loss of animals found negligent in our stewardship responsibility to regardless of their densities or nutritional state at the this unique resource. beginning of the die-off period. Once the snow and ice melts, the carrying capacity of the range instantly experiences a many-fold increase—the absolute for¬ Acknowledgments age supply is still there and is once again readily We thank the 1961 aerial survey crew, J. S. Tener, C. R. available to the animals. Harington, and D. C. Thomas of the Canadian Wildlife Therefore, caribou range in early winter that could Service (CWS), and pilots K. McLennan and W. Kellog, support several thousand animals under favorable Bradley Air Services, Ltd. for their pioneering work in snow and ice conditions can be rapidly reduced to one obtaining the 'benchmark' data set for Peary caribou on that will support only several hundred or fewer ani¬ the Queen Elizabeth Islands (QEI). Had the first system¬ mals. Then, following snowmelt, it will return imme¬ atic aerial survey of Peary caribou on the QEI been delayed diately to a relatively favorable state. Such capricious to the mid-1970s, our understanding of Peary caribou range conditions will not allow a predictable fixed numbers, distribution, and changes over time would be sustainable level of annual harvest of any appreciable highly distorted and dangerously misleading. We are size for more than a short series of years, if that. The grateful to the CWS, of Environment Canada (EC) and lack of adequate periodic monitoring programs and the Polar Continental Shelf Project of Natural Resources the uncertainty of 'climate change' leave the fate of Canada for supporting most of the studies of Peary caribou Peary caribou on the QEI unpredictable. Intervals of on the QEI. We also thank Agency of EC > 5 years between estimates of caribou on the QEI and the Department of Resources, Wildlife and Economic could run the risk of merely documenting serious Development (RWED) of the Government of Northwest declines after the fact, especially if continual annual Territories for their assistance over the years and, espe-

Rangifer, Special Issue No. 16, 2005 73 cially, A. Gunn and J. Dragon, RWED, for carrying out Johnson, D. H. 1999. The insignificance of statistical sig• the 1997 aerial survey of the western QEI. We thank J. nificance testing. — J. Wildl. Manage. 63: 763—772. S. Tener (past Director-General CWS), Ottawa, Ontario; Kelsall, J. P. 1968. The migratory barren-ground caribou of D. C. Thomas, CWS; and two anonymous reviewers for Canada. — Can. Wildl. Serv. Monogr. 3, Queen's Printer, critically reading the manuscript and making helpful com¬ Ottawa, Ontario, Canada, pp. 350 -384. ments and suggestions. We also thank G. Weiss, CWS, Krebs, C. J. 1999. Ecological Methodology, 2nd ed. Addi- EC, for providing Digital Elevation Model analysis and son—Wesley Longman, Inc., New York, NY, USA. map contours. Macpherson, A. H. 1961. On the abundance and distri• bution of certain mammals in the western Canadian Arctic Islands in 1958-59. — Arctic Circular 14: 1—17. References Manseau., M, Dick, L., Lyons, N., St.-Pierre. C. & Anderson, D. R., Link, W. A., Johnson, D. H. & Burn- Wood, J. 2004. Ecological history of Peary caribou and ham, K. P. 2001. Suggestions for presenting the results muskox on northern Ellesmere Island, ca. 4300 to pres• of data analyses. — J. Wildl. Manage. 65 : 373—378. ent. — Research Links 12 (1): 1, 4—8. Behnke, R. H. 2000. Equilibrium and non-equilibrium Maxwell, B. 1981. Climatic regions of the Canadian Arc• models of livestock population dynamics in pastoral tic Islands. — Arctic 34: 225—240. Africa: Their relevance to Arctic grazing systems. — Maxwell, B. 1997. Responding to global climate change Rangifer 20 (2-3): 141—152. in Canada's Arctic. — Environment Canada, Downsview, Bliss, L. C. 1990. High Arctic Ecosystems: how they Ontario, Canada, pp. 1—82. develop and are maintained. — In: Harington, C. R. Miller, F. L. 1987a. Peary caribou and muskoxen on (ed.). Canada's missing dimension: science and history inBathurst , Alexander, Marc, Massey, Vanier, Cameron, the Canadian Arctic Islands. Can. Mus. Nature, Ottawa, Helena, Lougheed, and Edmund Walker islands, North• Ontario, Canada, pp. 350—384. west Territories, July 1985. — Can. Wildl. Serv. Tech. Rep. Case, R. & Ellsworth, T. 1991. Distribution and abun¬ Ser. 20: 1—45. dance of muskoxen and Peary caribou on southern Elles- Miller, F. L. 1987b. Peary caribou and muskoxen on mere Island, NWT, July 1989. — , Eglinton Island, and Emerald Ile, Dep. Renew. Resourc., Manuscript Rep. 41: 1—23. Northwest Territories, July 1986. — Can. Wildl. Serv. Caughley, G. & Gunn, A. 1993. Dynamics of large her• Tech. Rep. Ser. 29: 1—65. bivores in deserts: kangaroos and caribou. — Oikos 67: Miller, F. L. 1988. Peary caribou and muskoxen on Mel¬ 47—55. ville and Byam Martin islands, Emerald Ile, Northwest Edlund, S. 1990. Bioclimatic zones in the Canadian Arc• Territories, July 1987. — Can. Wildl. Serv. Tech. Rep. Ser. tic Archipelago. — In: Harington, C. R. (ed.). Canada's 37: 1—58. missing dimension: science and history in the Canadian ArcticMiller , F. L. 1989. Reevaluation of the status of Peary cari¬ Islands. Can. Mus. Nature, Ottawa, Ontario, Canada, bou and muskox populations within the Bathurst Island pp. 421—441. complex, Northwest Territories, July 1988. — Can. Edlund, S. & Alt, B. 1989. Regional congruence of Wildl. Serv. Tech. Rep. Ser. 78: 1—86. vegetation and summer climate patterns in the Queen Miller, F. L. 1990a. Inter-island movements of Peary Elizabeth Islands, Northwest Territories, Canada. — Arc• caribou: a review and appraisement of their ecological tic 42: 3—23. importance. — In: Harington, C. R. (ed.). Canada's miss• Gauthier, L. 1996. Observations of wildlife on Ellesmere ing dimension: Science and history in the Canadian Arctic and Axel Heiberg islands between June 12—21, 1995. Islands. Can. Mus. Nature, Ottawa, Ontario, Canada, — Northwest Territories Dep. Renew. Resourc. Manuscriptpp . 608—632. Rep. 86: 1—28. Miller, F. L. 1990b. Peary caribou status report. Environ¬ Gunn, A. & Dragon, J. 2002. Peary caribou and mus- ment Canada report prepared for the Committee on the kox abundance and distribution on the western Queen Status of endangered Wildlife in Canada, Edmonton, Elizabeth Islands Northwest Territories and Alberta, Canada, pp. 1—64. June—July 1997. — Northwest Territories Dep. Resourc.,Miller , F. L. 1995a. Inter-island water crossings by Peary Wildl. and Economic Develop. File Rep. 130: 1—93. caribou, south-central Queen Elizabeth Islands. — Arctic Gunn, A., Miller, F. L. & Nishi, J. 2000. Status of 48: 8—12. endangered and threatened caribou on Canada's Arctic Miller, F. L. 1995b. Peary caribou conservation stud¬ Islands. — Rangifer Special Issue No. 12: 39—50. ies, Bathurst Island complex, Northwest Territories, Gunn, A., Miller, F. L. & Thomas, D. C. 1981. The cur¬ July—August 1993. — Can. Wildl. Serv. Tech. Rep. Ser. rent status and future of Peary caribou (Rangifer tarandus 230: 1—76. pearyi) on the Arctic Islands of Canada. — Biol. Conserv.Miller , F. L. 1997. Peary caribou conservation studies, 19(1980—81): 283—296. Bathurst Island complex, Northwest Territories, April—

74 Rangifer, Special Issue No. 16, 2005 August 1994 and June—July 1995. — Can. Wildl. Serv. western Queen Elizabeth Islands, Arctic Canada. — Can. Tech. Rep. Ser. 295: 1—155. J. Zool. 55: 1029—1037. Miller, F. L. 1998. Status of Peary caribou and muskox Parker, G. R., Thomas, D. C., Broughton, E. & Gray, populations within the Bathurst Island complex, south- D. R. 1975. Crashes of muskox and Peary caribou popu• central Queen Elizabeth Islands, Northwest Territories, lations in 1973—74 in the Parry Islands, Arctic Canada. July 1996. — Can. Wildl. Serv. Tech. Rep. Ser. 317: 1—147. — Can. Wildl. Serv. Prog. Notes 56: 1—10. Miller, F. L. 2002. Multi-island seasonal home use by Parry, W. E. 1821. Journal of a voyage for the discovery of a two Peary caribou, Canadian High Arctic, Nunavut, from the Atlantic to the Pacific; performed 1993—94. — Arctic 55: 133—142. in the years 1819—20, in His Majesty's ships Hecla and Miller, F. L. & Barry, S. J. 2003. Single-island home use Griper, under the Orders of ...with by four female Peary caribou, Canadian High Arctic an appendix containing the scientific and other observations. Islands, Nunavut, 1993—94. — Rangifer Special Issue John Murray, London, UK. No. 14: 267—281. Riewe, R. R. 1973. Final report on a survey of ungulate Miller, F. L. & Gunn, A. 2003a. Catastrophic die-off of populations on the Bjorne Peninsula, Ellesmere Island. Peary caribou on the western Queen Elizabeth Islands, Determination of numbers and distribution and assessment of Canadian High Arctic. — Arctic 56: 381—390. the effects of seismic activities on the behavoiur of these popu• Miller, F. L. & Gunn, A. 2003b. Status, population fluc¬ lations. Univ. Manitoba, Winnipeg, for Dep. Indian tuations and ecological relationships of Peary caribou and Northern Affairs, Ottawa, Ontario, Canada, pp. on the Queen Elizabeth Islands: Implications for their 1—59. survival. — Rangifer Special Issue No. 14: 213—226. Steidl, R. J., DeStefano, S. & Matter, W. J. 2000. On Miller, F. L., Russell, R. H. & Gunn, A. 1977a. Dis¬ increasing the quality, reliability, and rigor of wildlife tributions, movements and numbers of Peary caribou science. — Wildl. Soc. Bull. 28: 518—521. and muskoxen on western Queen Elizabeth Islands, Tener, J. S. 1963. Queen Elizabeth Islands game survey, Northwest Territories, 1972—74. — Can. Wildl. Serv. Rep. 1961. — Can. Wildl. Serv. Occas. Pap. 4: 1—50. Ser. 40: 1—55. Yoccoz, N. G. 1991. Use, overuse, and misuse of signifi• Miller, F. L., Russell, R. H. & Gunn, A. 1977b. Interisland cance tests in evolutionary biology and ecology. — Bull. movements of Peary caribou (Rangifer tarandus pearyi) on Ecological Soc. America 72: 106—111.

Rangifer, Special Issue No. 16, 2005 75